CN110703604B - Exoskeleton dynamic model parameter identification method and exoskeleton device - Google Patents

Exoskeleton dynamic model parameter identification method and exoskeleton device Download PDF

Info

Publication number
CN110703604B
CN110703604B CN201911034906.4A CN201911034906A CN110703604B CN 110703604 B CN110703604 B CN 110703604B CN 201911034906 A CN201911034906 A CN 201911034906A CN 110703604 B CN110703604 B CN 110703604B
Authority
CN
China
Prior art keywords
exoskeleton
data set
sampling
joint
connecting rod
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201911034906.4A
Other languages
Chinese (zh)
Other versions
CN110703604A (en
Inventor
郭庆
陈振雷
刘干
石岩
许猛
蒋丹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201911034906.4A priority Critical patent/CN110703604B/en
Publication of CN110703604A publication Critical patent/CN110703604A/en
Application granted granted Critical
Publication of CN110703604B publication Critical patent/CN110703604B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses an exoskeleton dynamics model parameter identification method and an exoskeleton device, which are applied to the field of exoskeleton robots and aim to solve the problems that the existing model identification method often cannot meet the identification precision and error convergence speed which are actually required; then, designing an excitation track of the exoskeleton under specific physical constraint by adopting an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; the sampling regression data set and the sampling torque data set form a sampling data set; finally, carrying out unknown parameter identification on the dynamic model according to the sampling data set by adopting an NFO algorithm; and designing a robust controller according to the identified kinetic model parameters, and driving the exoskeleton device in real time based on the designed robust controller.

Description

Exoskeleton dynamic model parameter identification method and exoskeleton device
Technical Field
The invention belongs to the field of exoskeleton robots, and particularly relates to an unknown parameter identification method and a control technology of an exoskeleton robot.
Background
The lower limb exoskeleton is a wearable device which combines human intelligence and mechanical strength, and has wide application scenes in aspects of rescue and relief, medical rehabilitation, military operation and the like. The lower limb exoskeleton system is a servo control system taking a servo device as a control core, and mainly comprises an electric signal processing device and a power mechanism. Typical exoskeleton components are as follows: (1) given the elements. It can be a mechanical device, such as a connecting rod, etc., providing a displacement signal; or an electrical element, such as a potentiometer, for providing a voltage signal; (2) a detection element. The feedback circuit is used for detecting the actual output quantity of the actuator and converting the actual output quantity into a feedback signal. It can be a photoelectric sensor, an encoder, a three-dimensional force sensor, etc.; (3) and a comparison element. For comparing the command signal with the feedback signal and deriving an error signal. In practice, there is generally no specific comparison element, but rather a structural element is part of the job; (4) and an amplifying and converting element. The error signal obtained by the comparison element is amplified and converted into an electrical or hydraulic signal (pressure, flow). It may be an electrical amplifier or the like; (5) and an execution element. Converting electric energy into mechanical energy, generating linear motion or rotary motion, and directly controlling the controlled object. Generally referred to as a motor or a hydraulic cylinder; (6) a controlled object. Refers to the load of the system, such as a workbench, etc.
The basic principle of the NFO (neighbor Field Optimization) algorithm is: and determining the number of individuals and an optimization ending condition on a search space according to the complexity of the optimization model, calculating adjacent dominant individuals and adjacent disadvantaged individuals of each individual, obtaining variation factors and variation vectors according to the number of the individuals and the optimization ending condition, and updating the individuals under the selected fitness function until the optimization ending condition is met.
With the increasingly expanded application of the exoskeleton robot in the engineering field, the requirement on the accuracy of exoskeleton parameter identification is higher and higher; the existing model identification method often cannot achieve the identification precision and the error convergence speed which are actually required, and the research of the model identification method with high precision and fast convergence is lacked.
Disclosure of Invention
In order to solve the technical problems, the invention provides an exoskeleton dynamics model parameter identification method based on an NFO algorithm, and the method is applied to a lower limb exoskeleton device, so that the accurate identification of the lower limb exoskeleton dynamics model parameters is realized, and the anti-interference capability of the lower limb exoskeleton device is improved by adopting the robust controller of the method.
One of the technical schemes adopted by the invention is as follows: an exoskeleton dynamics model parameter identification method based on an NFO algorithm comprises the following steps:
s1, establishing an exoskeleton dynamic model containing unknown parameters, and converting the exoskeleton dynamic model into a linear form;
s2, designing an excitation track of the exoskeleton under specific physical constraint by adopting an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; the sampling regression data set and the sampling torque data set form a sampling data set;
and S3, carrying out unknown parameter identification on the dynamic model according to the sampling data set by adopting an NFO algorithm.
Further, step S1 employs a lagrangian method to build a kinetic model of the exoskeleton.
Further, the excitation trajectory of the exoskeleton under the specific physical constraint in step S2 is specifically: the exoskeleton joint angle, the exoskeleton joint angular velocity and the exoskeleton joint angular acceleration meet the excitation track of the exoskeleton of specific physical constraints.
Further, step S3 is specifically to optimize a fitness function corresponding to the linear form of the exoskeleton dynamics model constructed according to the sampling data set by using the NFO algorithm, so as to identify parameters of the exoskeleton dynamics model.
Still further, the fitness function employs a Huber function.
Further, when the exoskeleton is a 2-DOF lower extremity exoskeleton, the NFO algorithm is used in step S3 to identify unknown parameters of the dynamical model by optimizing the following fitness function:
Figure BDA0002251214520000021
Figure BDA0002251214520000022
wherein E is(i)Is the square error of the ith generation data,
Figure BDA0002251214520000023
is the torque error of the ith generation data,
Figure BDA0002251214520000024
phi denotes a matrix of unknown parameters,
Figure BDA0002251214520000025
which represents an estimate of the value of phi,
Figure BDA0002251214520000026
is a sampling regression matrix of the ith generation data,
Figure BDA0002251214520000027
for the moment estimation of the ith generation of data,
Figure BDA0002251214520000028
is that
Figure BDA0002251214520000029
The (j) th element of (a),1，2respectively designed moment tau1,τ2The error threshold value of (2 ═ c [, ]1,2]T。
The second technical scheme adopted by the invention is as follows: a robust controller is obtained by designing the parameters of the dynamic model obtained by identification according to the method.
Further, when the exoskeleton is a 2-DOF lower extremity exoskeleton, the process of designing a robust controller is:
firstly, acquiring a state space model of a lower limb exoskeleton;
Figure BDA0002251214520000031
wherein
Figure BDA0002251214520000032
Superscript T is a transposition operation, z1＝x1-xr，z2＝x2-α，xrObtained from human gait;
then, aiming at the obtained state space model and exoskeleton dynamics model parameters obtained by identification based on the NFO algorithm, designing a robust controller:
Figure BDA0002251214520000033
wherein,
Figure BDA0002251214520000034
derivative of α, K2∈R2×2Is positively determinedThe matrix is a matrix of a plurality of matrices,
Figure BDA0002251214520000035
to virtually control the quantity, K1∈R2×2In order to be a positive definite matrix,
Figure BDA0002251214520000036
the estimated values of the gravity matrix, the Coriolis matrix and the joint friction moment are respectively.
The third technical scheme adopted by the invention is as follows: an exoskeleton device is driven by the robust controller.
Further, when the exoskeleton device is a 2-DOF lower extremity exoskeleton device, comprising: two mechanical linkages, noted: the device comprises a first connecting rod, a second connecting rod, two driving motors, two three-dimensional force sensors, two photoelectric encoders, four photoelectric sensors and a power supply; the first connecting rod is used as a thigh arm, the second connecting rod is used as a shank arm, the head end of the first connecting rod is used as a hip joint, the hinge joint between the first connecting rod and the second connecting rod is used as a knee joint, the hip joint and the knee joint are respectively provided with a driving motor, and the thigh arm and the shank arm are respectively provided with a three-dimensional force sensor;
further comprising: the thigh arm and the shank arm are respectively provided with a photoelectric encoder, the thigh arm and the shank arm are respectively provided with two photoelectric sensors for detecting respective limit signals of the first connecting rod and the second connecting rod, and the thigh arm and the shank arm are respectively provided with two bandages for coupling the lower limb exoskeleton device with a human body;
the power supply supplies power for the driving motor, the three-dimensional force sensor, the photoelectric encoder and the photoelectric sensor.
The invention has the beneficial effects that: firstly, establishing a Lagrange dynamics model of the lower limb exoskeleton, and then obtaining a linear form of the lower limb exoskeleton according to a regression matrix and a parameter matrix; designing an excitation track of the lower limb exoskeleton under specific physical constraint based on an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; identifying unknown parameters of the model according to the sampling data set by adopting an NFO algorithm, and taking a Huber function as a fitness function; designing a robust controller according to the identified Lagrange model parameters; the method carries out parameter identification on the Lagrange model of the lower limb exoskeleton based on the NFO algorithm, realizes robust control on the Lagrange model, and improves the anti-interference capability of the system.
Drawings
FIG. 1 is a flow chart of a method for identifying exoskeleton dynamics model parameters based on the NFO algorithm according to the present invention;
fig. 2 is a mechanism schematic diagram of a 2-DOF lower extremity exoskeleton provided in an embodiment of the present invention.
Detailed Description
To facilitate understanding of the technical content of the present invention by those skilled in the art, the present invention will be further explained with reference to fig. 1-2.
In order to improve the accuracy of exoskeleton parameter identification, the invention provides an exoskeleton dynamic model parameter identification method based on an NFO algorithm, and as shown in FIG. 1, the method of the invention comprises the following steps:
s1, establishing an exoskeleton dynamic model containing unknown parameters, and converting the exoskeleton dynamic model into a linear form;
s2, designing an excitation track of the exoskeleton under specific physical constraint by adopting an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; the sampling regression data set and the sampling torque data set form a sampling data set;
and S3, carrying out unknown parameter identification on the dynamic model according to the sampling data set by adopting an NFO algorithm.
While the present embodiment is described with reference to a 2-dof (degree of freedom) lower extremity exoskeleton apparatus as shown in fig. 2, it should be noted by those skilled in the art that the present invention is not limited to a 2-degree-of-freedom lower extremity exoskeleton, but is also applicable to other degrees of freedom exoskeleton (including upper and lower extremities exoskeleton).
As shown in fig. 2, the 2-DOF lower extremity exoskeleton device comprises: two mechanical links, respectively noted: the device comprises a first connecting rod, a second connecting rod, two driving motors and two three-dimensional force sensors; wherein the first connecting rod is called thigh arm, and the second connecting rod is called shank arm; the head end of the first connecting rod is called a hip joint, and the articulation between the first connecting rod and the second connecting rod is called a knee joint; the hip joint and the knee joint are respectively provided with a driving motor for providing driving torque; the thigh arm and the shank arm are respectively provided with a three-dimensional force sensor for measuring human-computer interaction force;
in the embodiment, 2 24A direct current power supplies are arranged on the whole 2-DOF lower limb exoskeleton device and used for providing energy for the device, specifically supplying power for a driving motor, a three-dimensional force sensor, a photoelectric encoder and a photoelectric sensor; the first connecting rod and the second connecting rod are respectively provided with a photoelectric encoder for measuring the movement angle and the angular velocity of two joints, the first connecting rod and the second connecting rod are respectively provided with two photoelectric sensors for detecting the limiting signals of the two joints, and the first connecting rod and the second connecting rod are respectively provided with two bandages for coupling a human body and the exoskeleton. The direction of gravity is shown in FIG. 2.
In an alternative embodiment of the present invention, step S1 employs the lagrangian method to build a 2-DOF lower extremity exoskeleton dynamical model, which is represented as:
Figure BDA0002251214520000051
where θ is the joint angle of the exoskeleton, θ ∈ R2(R represents a real number set, the same applies hereinafter), τ is a motor-driven joint torque, τ ∈ R2，τdisIs a man-machine coupling moment, taudis＝JT(θ)Fdis，JT(θ) is the Jacobian matrix of the system, JT(θ)∈R2×2，FdisIs a coupling force between man and machine, Fdis∈R2；M(θ)∈R2×2、
Figure BDA0002251214520000052
G(θ)∈R2Representing the inertia matrix, coriolis matrix, and gravity matrix, respectively.
The exoskeleton with other degrees of freedom corresponds to the dynamic models with different dimensions, and the exoskeleton dynamic model parameters with other degrees of freedom can be identified according to the technical scheme of the invention.
M(θ)、
Figure BDA0002251214520000053
The expression of each of G (θ) is as follows:
Figure BDA0002251214520000054
Figure BDA0002251214520000055
Figure BDA0002251214520000056
wherein,
Figure BDA0002251214520000057
respectively represents the angular velocity and the angular acceleration of the exoskeleton hip joint,
Figure BDA0002251214520000058
respectively representing the angular velocity and the angular acceleration of the exoskeleton knee joint Ith,IshRespectively represent the inertia of the thigh arm and the shank arm of the exoskeleton, mth,mshRespectively representing the mass of the thigh arm and the forearm of the exoskeleton, athIndicates the length of the exoskeleton thigh arm, lth,lshRespectively showing the length from the hip joint to the mass center of the thigh arm, the length from the knee joint to the shank arm,
Figure BDA0002251214520000059
indicating joint friction.
In step S1, a parameter matrix phi and a regression matrix are used
Figure BDA00022512145200000510
Linearizing the Lagrange model into the following linear form, specifically:
Figure BDA00022512145200000511
wherein phi is an unknown parameter matrix whose elements are to be identified,
Figure BDA00022512145200000512
is a regression matrix;
the expression of the parameter matrix Φ is:
Φ＝[Φ(1)Φ(2)Φ(3)Φ(4)Φ(5)Φ(6)Φ(7)Φ(8)]T，
wherein,
Figure BDA0002251214520000061
Figure BDA0002251214520000062
Φ(5)＝k1,1，Φ(6)＝k1,2，Φ(7)＝k2,1，Φ(8)＝k2,2。
regression matrix
Figure BDA0002251214520000063
The expression of (a) is:
Figure BDA0002251214520000064
wherein,
Figure BDA0002251214520000065
Figure BDA0002251214520000066
Y(14)＝esinθ1，
Figure BDA0002251214520000067
Figure BDA0002251214520000068
Y(17)＝0，Y(18)＝0，Y(21)＝0，
Figure BDA0002251214520000069
Y(24)＝0，Y(25)＝0，Y(26)＝0，
Figure BDA00022512145200000610
in an alternative embodiment of the present invention, the step S2 designs the excitation trajectory into a fourier series form according to the property of the fourier series, as follows:
Figure BDA00022512145200000611
wherein, thetadDesired joint trajectory, θi,oIs an offset value of the joint locus, k is a frequency coefficient, kfIs the fundamental frequency, t is time, N is the period, ak,bkOptimization parameters required for the solution required by the NFO algorithm.
The specific implementation process of step S3 is to minimize the following fitness function by using the NFO algorithm:
Figure BDA00022512145200000612
where T is the specific physical constraints of a given lower extremity exoskeleton, including the rotational angle, angular velocity, angular acceleration of the exoskeleton,
Figure BDA00022512145200000613
in order to sample the regression matrix,
Figure BDA00022512145200000614
to represent
Figure BDA00022512145200000615
Condition number of (2).
In an alternative embodiment of the inventionStep S3 uses
Figure BDA00022512145200000616
And
Figure BDA00022512145200000617
forming a sampling data set obtained in the excitation experiment, and optimizing and estimating an unknown parameter matrix phi by using an NFO algorithm according to the linear form of the exoskeleton model obtained in the step S1 to obtain the unknown parameter matrix phi
Figure BDA0002251214520000071
To reduce the estimation error of the interference point pairs in the sampled data set
Figure BDA0002251214520000072
The following Huber function is adopted as a fitness function of the NFO algorithm to improve the robustness of system identification.
Figure BDA0002251214520000073
Figure BDA0002251214520000074
Wherein E is(i)Is the square error of the ith generation data,
Figure BDA0002251214520000075
is the torque error of the ith generation data,
Figure BDA0002251214520000076
for the moment estimation of the ith generation of data,
Figure BDA0002251214520000077
is that
Figure BDA0002251214520000078
The (j) th element of (a),1，2respectively designed moment tau1,τ2The error threshold value of (2 ═ c [, ]1,2]T。
According to the parameters identified by the exoskeleton Lagrange model parameter identification method based on the NFO algorithm, a robust controller is designed, and the method comprises the following steps:
a1, obtaining a state space model of the lower extremity exoskeleton, wherein the state space model comprises the following steps:
Figure BDA0002251214520000079
wherein,
Figure BDA00022512145200000710
z1＝x1-xr，z2＝x2-α，xrobtained from human gait.
A2, designing a robust controller for the state space model as follows:
Figure BDA00022512145200000711
wherein, the superscript T is the transposition operation,
Figure BDA00022512145200000712
derivative of α, K2∈R2×2In order to be a positive definite matrix,
Figure BDA00022512145200000713
to virtually control the quantity, K1∈R2×2In order to be a positive definite matrix,
Figure BDA00022512145200000714
estimated values of gravity matrix, Coriolis matrix, joint friction torque, Gn,Cn,Mn,τfnRespectively satisfy
Figure BDA00022512145200000715
The embodiment is based on the Lyapunov energy function, and proves the stability of the 2-DO lower limb exoskeleton device under the identified model parameters and the designed robust controller; the method specifically comprises the following steps:
set the Lyapuloff function as
Figure BDA0002251214520000081
The lyapuloff energy function is well known in the art and the present invention is not described in detail herein.
Aiming at the problem that the control performance and stability of an exoskeleton system are influenced by the existence of unknown parameters in an exoskeleton Lagrange model, the method adopts an NFO algorithm to identify the exoskeleton Lagrange model, and designs a robust controller by utilizing the identified model, thereby improving the control performance and stability of the lower limb exoskeleton with unknown parameters.
Firstly, establishing a Lagrange model of a 2-DOF lower limb exoskeleton, and then obtaining a linear form of the Lagrange model according to a regression matrix and a parameter matrix; designing an excitation track of the lower limb exoskeleton under specific physical constraint based on an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; identifying unknown parameters of the model according to the sampling data set by adopting an NFO algorithm, and taking a Huber function as a fitness function; designing a robust controller according to the identified Lagrange model parameters; based on the Lyapunov energy function, the stability of the exoskeleton system under the identified model parameters and the designed robust controller is proved; and finally, driving the 2-DOF lower limb exoskeleton in real time according to the designed robust controller.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims (8)

1. An exoskeleton dynamics model parameter identification method based on an NFO algorithm is characterized by comprising the following steps:
s1, establishing an exoskeleton dynamic model containing unknown parameters, and converting the exoskeleton dynamic model into a linear form;
s2, designing an excitation track of the exoskeleton under specific physical constraint by adopting an NFO algorithm, and performing an excitation experiment to obtain a sampling regression data set and a sampling torque data set; the sampling regression data set and the sampling torque data set form a sampling data set;
s3, carrying out unknown parameter identification on the dynamic model according to the sampling data set by adopting an NFO algorithm; specifically, a fitness function corresponding to a linear form of the exoskeleton dynamics model constructed according to the sampling data set is optimized by adopting an NFO algorithm, so that exoskeleton dynamics model parameters are distinguished;
when the exoskeleton is a 2-DOF lower extremity exoskeleton, the NFO algorithm is used to identify unknown parameters of the dynamical model by optimizing the following fitness function in step S3:
Figure FDA0002484048960000011
Figure FDA0002484048960000012
wherein E is(i)Is the square error of the ith generation data,
Figure FDA0002484048960000013
is the torque error of the ith generation data,
Figure FDA0002484048960000014
phi denotes a matrix of unknown parameters,
Figure FDA0002484048960000015
which represents an estimate of the value of phi,
Figure FDA0002484048960000016
is a sampling regression matrix of the ith generation data,
Figure FDA0002484048960000017
for the moment estimation of the ith generation of data,
Figure FDA0002484048960000018
is that
Figure FDA0002484048960000019
The (j) th element of (a),1，2respectively designed moment tau1,τ2The error threshold value of (2 ═ c [, ]1,2]T。
2. The method for identifying parameters of an exoskeleton of claim 1, wherein step S1 is to establish a kinetic model of the exoskeleton using lagrangian method.
3. The method of claim 2, wherein the excitation trajectory of the exoskeleton under the specific physical constraints in step S2 is specifically: the exoskeleton joint angle, the exoskeleton joint angular velocity and the exoskeleton joint angular acceleration meet the excitation track of the exoskeleton of specific physical constraints.
4. The method of claim 3, wherein the fitness function is a Huber function.
5. A robust controller, characterized by being designed using kinetic model parameters identified by the method of any of claims 1-4.
6. The robust controller of claim 5, wherein when said exoskeleton is a 2-DOF lower extremity exoskeleton, said process of designing a robust controller is:
firstly, acquiring a state space model of a lower limb exoskeleton;
Figure FDA0002484048960000021
wherein,
Figure FDA0002484048960000022
Figure FDA0002484048960000023
shows the angular velocity of the exoskeleton hip joint,
Figure FDA0002484048960000024
Representing the angular acceleration of the exoskeleton hip joint,
Figure FDA0002484048960000025
the angular velocity of the exoskeleton knee joint is shown,
Figure FDA0002484048960000026
representing angular acceleration of the exoskeleton knee joint, the superscript T being the transposition operation, z1＝x1-xr，z2＝x2-α，xrObtained from human gait;
then, aiming at the obtained state space model and exoskeleton dynamics model parameters obtained by identification based on the NFO algorithm, designing a robust controller:
Figure FDA0002484048960000027
wherein,
Figure FDA0002484048960000028
derivative of α, K2∈R2×2In order to be a positive definite matrix,
Figure FDA0002484048960000029
to virtually control the quantity, K1∈R2×2Is a positive definite matrix, tau is the motor-driven joint moment, taudisIs man-machine coupling moment, M represents inertia matrix, C represents Coriolis matrix, G represents gravity matrix, and taufThe joint friction torque is represented by the torque of the joint,
Figure FDA00024840489600000210
the estimated values of the gravity matrix, the Coriolis matrix and the joint friction moment are respectively.
7. An exoskeleton device driven by the robust controller of claim 6.
8. The exoskeleton device of claim 7, when the exoskeleton device is a 2-DOF lower extremity exoskeleton device, comprising: two mechanical linkages, noted: the device comprises a first connecting rod, a second connecting rod, two driving motors, two three-dimensional force sensors, two photoelectric encoders, four photoelectric sensors and a power supply; the first connecting rod is used as a thigh arm, the second connecting rod is used as a shank arm, the head end of the first connecting rod is used as a hip joint, the hinge joint between the first connecting rod and the second connecting rod is used as a knee joint, the hip joint and the knee joint are respectively provided with a driving motor, and the thigh arm and the shank arm are respectively provided with a three-dimensional force sensor;
further comprising: the thigh arm and the shank arm are respectively provided with a photoelectric encoder, the thigh arm and the shank arm are respectively provided with two photoelectric sensors for detecting respective limit signals of the first connecting rod and the second connecting rod, and the thigh arm and the shank arm are respectively provided with two bandages for coupling the lower limb exoskeleton device with a human body;
the power supply supplies power for the driving motor, the three-dimensional force sensor, the photoelectric encoder and the photoelectric sensor.
CN201911034906.4A 2019-10-29 2019-10-29 Exoskeleton dynamic model parameter identification method and exoskeleton device Expired - Fee Related CN110703604B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911034906.4A CN110703604B (en) 2019-10-29 2019-10-29 Exoskeleton dynamic model parameter identification method and exoskeleton device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911034906.4A CN110703604B (en) 2019-10-29 2019-10-29 Exoskeleton dynamic model parameter identification method and exoskeleton device

Publications (2)

Publication Number Publication Date
CN110703604A CN110703604A (en) 2020-01-17
CN110703604B true CN110703604B (en) 2020-07-28

Family

ID=69202610

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911034906.4A Expired - Fee Related CN110703604B (en) 2019-10-29 2019-10-29 Exoskeleton dynamic model parameter identification method and exoskeleton device

Country Status (1)

Country Link
CN (1) CN110703604B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111775142B (en) * 2020-08-12 2021-12-14 电子科技大学 Model identification and self-adaptive control method for hydraulic mechanical arm
CN113408066B (en) * 2021-06-15 2023-12-22 军事科学院系统工程研究院军需工程技术研究所 Method and device for identifying human-machine posture deviation of knee joint exoskeleton based on subspace
CN113855477B (en) * 2021-09-07 2022-07-19 燕山大学 Layered control method for lower limb exoskeleton robot
CN114721258B (en) * 2022-02-21 2023-03-10 电子科技大学 Lower limb exoskeleton backstepping control method based on nonlinear extended state observer
CN115524970B (en) * 2022-09-28 2024-08-06 中国南方电网有限责任公司超高压输电公司昆明局 Exoskeleton local pressure improvement method applied to extra-high voltage direct current high-altitude operation
CN116922396B (en) * 2023-09-07 2023-11-17 贵州航天控制技术有限公司 Assistance control method for exoskeleton system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101221437A (en) * 2008-01-24 2008-07-16 上海交通大学 Industrial production full process optimizing and controlling method in network information interchange mode
CN109375516A (en) * 2018-12-07 2019-02-22 国家康复辅具研究中心 A kind of human innertial parameter's discrimination method of binding kinetics constraint condition

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ATE531313T1 (en) * 2009-04-07 2011-11-15 Syco Di Hedvig Haberl & C S A S SYSTEM FOR CONTROLLING A HAPTIC EXOSKELETON DEVICE FOR REHABILITATION PURPOSES AND CORRESPONDING HAPTIC EXOSKELETON DEVICE
CN104575021B (en) * 2014-12-17 2016-06-15 浙江工业大学 Distributed model predictive control method based on neighborhood Optimizing City road network system
CN105108761B (en) * 2015-08-14 2017-05-24 浙江大学 Reduced-order adaptive robust cascading force control method for single-joint powered exoskeleton
CN105116728B (en) * 2015-08-14 2018-02-27 安徽三联机器人科技有限公司 A kind of method of simple joint assistance exoskeleton inverting ADAPTIVE ROBUST power control
CN105796286B (en) * 2016-02-23 2018-04-06 东南大学 Use the lower limb exoskeleton robot control method of air bag sensor
CN107498562A (en) * 2017-04-21 2017-12-22 浙江工业大学 Six-degree-of-freedom mechanical arm dynamics model identification method
CN107397649B (en) * 2017-08-10 2020-08-14 燕山大学 Upper limb exoskeleton movement intention identification method based on radial basis function neural network

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101221437A (en) * 2008-01-24 2008-07-16 上海交通大学 Industrial production full process optimizing and controlling method in network information interchange mode
CN109375516A (en) * 2018-12-07 2019-02-22 国家康复辅具研究中心 A kind of human innertial parameter's discrimination method of binding kinetics constraint condition

Also Published As

Publication number Publication date
CN110703604A (en) 2020-01-17

Similar Documents

Publication Publication Date Title
CN110703604B (en) Exoskeleton dynamic model parameter identification method and exoskeleton device
Li et al. Model-free control for continuum robots based on an adaptive Kalman filter
CN108656112B (en) Mechanical arm zero-force control experiment system for direct teaching
Li et al. Asymmetric bimanual control of dual-arm exoskeletons for human-cooperative manipulations
Wang et al. Dynamic control of multisection three-dimensional continuum manipulators based on virtual discrete-jointed robot models
Li et al. Adaptive human–robot interaction control for robots driven by series elastic actuators
Li et al. Vision-based robotic manipulation of flexible PCBs
CN112276944A (en) Man-machine cooperation system control method based on intention recognition
Kormushev et al. Kinematic-free position control of a 2-dof planar robot arm
Solazzi et al. An interaction torque control improving human force estimation of the rehab-exos exoskeleton
Saków et al. Self-sensing teleoperation system based on 1-dof pneumatic manipulator
Sadun et al. An overview of active compliance control for a robotic hand
Yu et al. Position-based visual servo control of dual robotic arms with unknown kinematic models: A cerebellum-inspired approach
Aksman et al. Force estimation based compliance control of harmonically driven manipulators
Gao et al. Design and analysis of underactuated robotic gripper with adaptive fingers for objects grasping tasks
Hu et al. Motion and force control of coordinated robots during constrained motion tasks
Zhang et al. Learning-based tracking control of soft robots
TWI781708B (en) Learning apparatus, learning method, learning program, control apparatus, control method, and control program
Minyong et al. Hybrid impedance and force control for massage system by using humanoid multi-fingered robot hand
Zhang et al. Adaptive Neural Network Control for Exoskeleton Motion Rehabilitation Robot With Disturbances and Uncertain Parameters
Chen et al. Pneumatic actuation-based bidirectional modules with variable stiffness and closed-loop position control
Farooq et al. Improved hybrid position/force controller design of a flexible robot manipulator using a sliding observer
Hua et al. A force/position hybrid controller for rehabilitation robot
Wang et al. Decentralized Position/torque Control of Modular Robot Manipulators via Interaction Torque Estimation-based Human Motion Intention Identification
Lei et al. Vision-based position/impedance control for robotic assembly task

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200728

Termination date: 20211029